From the Guidelines
AI tools for corneal assessment and diagnosis are increasingly being integrated into ophthalmology practice, with the most recent and highest quality study suggesting that these tools can improve diagnostic accuracy and patient outcomes. The most widely used AI applications for cornea include automated analysis of corneal topography and tomography, which help detect keratoconus and other ectatic disorders earlier than traditional methods 1. Specific systems like the Pentacam AXL Wave with its TBI (Tomographic Biomechanical Index) and the Oculus Corvis ST with Vinciguerra Screening Report use machine learning algorithms to identify subtle corneal changes. For corneal imaging, AI tools can analyze confocal microscopy and OCT images to quantify endothelial cell density, identify corneal nerve damage in conditions like dry eye disease, and assess corneal thickness with high precision. Some key points to consider when evaluating AI tools for corneal assessment include:
- The ability of the AI system to detect different types of ocular images and gradeability of images 1
- The operational workflow of the AI system, including image preprocessing, analysis, and generation of diagnosis 1
- The clinical gap that the proposed AI algorithm may bridge, including its application on specific target populations and clinical settings 1 In surgical planning, AI systems assist in optimizing corneal transplantation procedures by helping surgeons determine graft size and position. These tools improve diagnostic accuracy by eliminating subjective interpretation variability and can detect patterns invisible to the human eye. The integration of these AI systems into clinical workflows allows for earlier intervention in progressive corneal diseases, potentially preserving vision that might otherwise be lost to conditions like advanced keratoconus requiring corneal transplantation. Overall, the use of AI tools for corneal assessment and diagnosis has the potential to significantly improve patient outcomes, and their integration into clinical practice is highly recommended.
From the Research
AI Tools for Cornea
There are no direct research papers on AI tools for cornea in the provided evidence. However, the studies discuss various corneal diseases and their diagnosis, treatment, and management.
Corneal Diseases and AI
- Keratoconus (KC) and Fuchs' endothelial corneal dystrophy (FECD) are two corneal diseases that have been studied extensively 2, 3, 4.
- The comorbidity of KC and FECD has been reported in several cases, with a possible genetic link between the two diseases 3.
- The diagnosis and treatment of corneal injuries, including non-infectious corneal injuries, have been reviewed, with various pharmacological agents and therapeutic medications discussed 5.
- Fuchs endothelial corneal dystrophy (FECD) is a common corneal dystrophy that frequently results in vision loss, with definitive treatment being corneal transplantation 6.
Potential Applications of AI in Corneal Diseases
- AI can potentially be used to develop diagnostic tools for corneal diseases, such as KC and FECD, by analyzing images and data from patients 2, 3, 4.
- AI can also be used to develop personalized treatment plans for patients with corneal diseases, taking into account their individual characteristics and needs 5.
- Additionally, AI can be used to analyze data on corneal diseases, such as FECD, to better understand the disease's clinical features, pathophysiology, and genetics 6.